BioContainers Web

The main page of BioContainers and the registry, served as a fully static site on GitHub Pages — no backend, no database. The tool catalog is regenerated from upstream by a pipeline that runs in GitHub Actions and is committed into the repo as static JSON; search runs entirely in the browser.

Repository layout

• pipeline/ — Python CLI that regenerates the static catalog from two upstream families:

  • bioconda recipes + quay.io/biocontainers container tags
  • BioContainers/containers Dockerfiles + DockerHub biocontainers tags

  It writes the data contract into data/: search-index.json, tools/<id>.json, facets.json, stats.json. See docs/ for the design spec and plans.

• web/ — the Vue 3 + Vite single-page app. It reads only the static data/ files and searches client-side with MiniSearch. Three screens: Home (stats), Registry (search + facets), Tool detail (versions + container pull commands).

• data/ — the generated catalog, committed to the repo and served alongside the app.

Workflows

• .github/workflows/data-build.yml — weekly (and manual) run of the pipeline; commits an updated data/ to the default branch.
• .github/workflows/deploy.yml — on push, builds web/ and publishes it (with data/) to GitHub Pages at biocontainers.pro.

Local development

Build a small sample catalog and run the app against it:

# 1. Generate a sample data/ catalog (needs Python 3.11+)
cd pipeline
python -m venv .venv && .venv/bin/pip install -e ".[dev]"
.venv/bin/python -m biocontainers_pipeline.cli build --out ../data --only quay --limit 60

# 2. Run the web app (needs Node 20+); web/public/data symlinks to ../../data
cd ../web
npm install
npm run dev

Tests: cd pipeline && .venv/bin/python -m pytest and cd web && npm test.

Contributing

Open a PR against the default branch; merges deploy automatically to https://biocontainers.pro